DESCRIPTION (Applicant's Description): High-resolution imaging allows improved understanding of retinal disease and pathological conditions, particularly in prognosis of malignant choroidal melanoma, age-related macular degeneration, diabetic retinopathy, ocular complications of AIDS and glaucoma. The eye can only be regarded as a diffraction-limited optical system up to a pupil diameter of 3 mm. Standard retinal imaging technology uses a 3 mm imaging aperture to avoid the higher order aberrations of the outer cornea. However, if these higher order aberrations can be compensated, diffraction-limited imaging at 7 or 8 mm pupil diameter can be achieved. The numerical aperture of the diffraction-limited eye at 7 or 8 mm pupil will allow to visualize retinal detail that was previously not achievable in ophthalmology. To achieve this goal we plan to develop a wavefront sensor that will allow us to measure the existing aberrations of the measured eye. We will test it in an eye model, in animal eyes and in human eyes. The wavefront sensor will be attached to a fundus camera and scanning laser ophthalmoscopes. The second aim is to use an adaptive wavefront compensator to correct the aberrations in a feed-back loop setup. We will test the complete system of wavefront sensor and wavefront compensator in an eye model, in animal eyes and in human eyes. The wavefront compensator will be attached to a fundus camera and scanning laser ophthalmoscopes. The third aim is to used digital image processing to correct for the portion of the residual aberrations that could not be compensated with the adaptive wavefront compensator due to mechanical considerations. Our group has previous developed a wavefront sensor and wavefront compensator based on a micromachined membrane deformable mirror. Due to the limited number of electrodes, slow speed of our computer system and the mechanical stiffness of the membrane we were not able to completely correct all measured aberrations. Our preliminary work shows that the system is capable of allowing wavefront correction. In this study we plan to improve the acquisition speed of our wavefront sensor and wavefront compensator to allow rapid wavefront compensation. Our preliminary results have shown that even the best aberration compensation still suffers from residual wave aberrations. Since we can measure and characterize these aberrations, we can develop digital inverse filter based on our experience in image reconstruction to correct the acquired images.